It’s notable that just the English “implementation” of FizzBuzz here is longer and more ambiguous than the naive Python implementation, never mind the boilerplate (which itself is also longer than the Python).
The explosion of frameworks and YAML tools the author describes can be attributed to the fact that English is an extremely poor language for program specification, and requires all kinds of guardrails and annotation to accomplish the same specificity as a typical computer program.
I thought for a second you were talking about Logos Lang! The language I’m working on that we believe will be humanities last language. Best of luck, and good to have competition. How’s your FFI support? ;) https://www.logicaffeine.com/
I've recently hand the 'pleasure' of dealing with legal issues. Google ai was giving me lots of useful sounding advice but you ask for case law, and then Google itself can't find it, and the ai in that search is guessing that you've made a mistake, because the citations just don't exist.
But then you have programming where it appears to be scarily good.
So is this a case of programmers program so their Ais are tested to be good at programming? Or is this cognitive dissonance on our part?
To tie it in to this post. Is it a good thing to have the ai generate the code and data? It assumes an understanding that isn't really there.
I used Opus 4.6. In my defense, in addition to wanting to try a live version of this because I think it's super cool, I was also academically curious if we've gotten to the point where we have one-shot blog post -> compiler pipelines working.
> Metanarratively, the replicants weren't the problem. The problem was the people that made them. The people that gave them the ability to think. The ability to feel. The ability to understand and emphathize. The problem was the people that gave them the ability to enjoy life and then hit them with a temporal Sword of Damocles overhead because those replicants were fundamentally disposable.
> In Blade Runner, the true horror was not the technology. The technology worked fine. The horror was the deployment and the societal implications around making people disposable. I wonder what underclass of people like that exists today.
That underclass is non human animals. We breed tens of billions of them into existence every year. They have the ability to feel, think, enjoy life, understand and even likely empathize. They don’t want to suffer and they want to live. And they are enslaved, tortured, and then brutally killed all for a few minutes of pleasure, when alternatives exist in abundance.
not sure if that's a direct comparison to bladerunner. I've not watched/read it but just skimming the wikipedia article, replicants seem to have human-level sapience; I think the comparison was rather to oppressed human underclasses.
regardless, what you say is true and that's an interesting comparison.
Why make markdown a typed programming language instead of using protobuf/text proto?
LLMs already are very familiar with it, there are 100s of protoc plugin to generate code from it, and it's less verbose and more token efficient + testable with protovalidate
Give me a fucking break. You’re treating Blade Runner like some high‑falutin moral fable while the actual underclass is sitting in plain sight, exploited by systems you refuse to even look up. Modern slavery isn’t just a dystopian backdrop; over fifty million people live in some form of forced labor or forced marriage right now, and that number has been climbing, not stabilizing. Tens of millions are stuck in forced labor for companies and states, generating hundreds of billions in illegal profits off human coercion every single year. Countries like India, China, North Korea, Pakistan, Russia, Indonesia, Nigeria, Bangladesh, Turkey, and the United States together account for well over half of all victims, and some of the worst spikes are happening in Iraq, Venezuela, and countries already drowning in climate‑related disasters.
If you want a real starting point that actually hits you in the chest rather than in the feel‑good aesthetic cortex, look at wage theft in the United States. Wage theft outright steals about fifty billion dollars from workers each year. That’s over a hundred times what robberies cost in terms of direct financial loss, and it’s happening to the very people who can least afford it. At least four million workers are paid below the minimum wage illegally every year, each losing roughly three thousand dollars on average, which adds up to over thirteen billion dollars in stolen wages just from that one category. That doesn’t even include unpaid overtime, not being paid for full hours, being misclassified as “independent contractors,” or being forced to work “off the clock” with no record and no recourse.
Low‑wage workers in service jobs, domestic work, farming, and precarious gig‑adjacent roles are the ones getting hammered hardest, with surveys in several cities showing around half of low‑wage workers reporting at least one serious wage violation each year. You can’t get poorer than having time stolen from you, and that’s exactly what’s happening on a massive scale. The feds barely lift a finger: the agency meant to enforce basic wage rules has fewer than six hundred and fifty investigators overseeing more than a hundred and sixty‑five million workers, which is one investigator for roughly three hundred thousand people. In the most recent year they recovered a bit over a quarter billion dollars in back pay, but that’s less than one percent of the estimated fifty billion stolen from workers in that same year. In plain English, the state is letting something like ninety‑five percent or more of wage theft go utterly unpunished.
So no, Blade Runner is not your wake‑up call. The film is a mood board for people who want to pretend they’re deep while keeping their eyes closed to the real indices of exploitation happening right now, right here—the forty million people trapped in modern slavery worldwide, the tens of millions in forced labor, and the fifty billion dollars stolen from workers in your own backyard every year. Those are the numbers that should disgust you, not some neon‑soaked metaphor about replicants. If you can’t feel rage in front of that, you’re not woke; you’re just playing dress‑up in someone else’s hellscape.
It’s notable that just the English “implementation” of FizzBuzz here is longer and more ambiguous than the naive Python implementation, never mind the boilerplate (which itself is also longer than the Python).
The explosion of frameworks and YAML tools the author describes can be attributed to the fact that English is an extremely poor language for program specification, and requires all kinds of guardrails and annotation to accomplish the same specificity as a typical computer program.
> is longer and more ambiguous than the naive Python implementation
write it in lojban instead, ez
I thought for a second you were talking about Logos Lang! The language I’m working on that we believe will be humanities last language. Best of luck, and good to have competition. How’s your FFI support? ;) https://www.logicaffeine.com/
Markdown with front matter? Pppfff.
Challenge: I'll only switch to this brave new world of "LLM compilers" if the "spec language" is XML.
=)
Make it so, HN brain trust!
I've recently hand the 'pleasure' of dealing with legal issues. Google ai was giving me lots of useful sounding advice but you ask for case law, and then Google itself can't find it, and the ai in that search is guessing that you've made a mistake, because the citations just don't exist.
But then you have programming where it appears to be scarily good.
So is this a case of programmers program so their Ais are tested to be good at programming? Or is this cognitive dissonance on our part?
To tie it in to this post. Is it a good thing to have the ai generate the code and data? It assumes an understanding that isn't really there.
If anyone is curious I implemented a prototype of this at https://github.com/avaer/markdownlang
I used Opus 4.6. In my defense, in addition to wanting to try a live version of this because I think it's super cool, I was also academically curious if we've gotten to the point where we have one-shot blog post -> compiler pipelines working.
Turns out, the answer is yes.
> Metanarratively, the replicants weren't the problem. The problem was the people that made them. The people that gave them the ability to think. The ability to feel. The ability to understand and emphathize. The problem was the people that gave them the ability to enjoy life and then hit them with a temporal Sword of Damocles overhead because those replicants were fundamentally disposable.
> In Blade Runner, the true horror was not the technology. The technology worked fine. The horror was the deployment and the societal implications around making people disposable. I wonder what underclass of people like that exists today.
That underclass is non human animals. We breed tens of billions of them into existence every year. They have the ability to feel, think, enjoy life, understand and even likely empathize. They don’t want to suffer and they want to live. And they are enslaved, tortured, and then brutally killed all for a few minutes of pleasure, when alternatives exist in abundance.
not sure if that's a direct comparison to bladerunner. I've not watched/read it but just skimming the wikipedia article, replicants seem to have human-level sapience; I think the comparison was rather to oppressed human underclasses.
regardless, what you say is true and that's an interesting comparison.
I wrote something similar: https://intentcode.dev
Why make markdown a typed programming language instead of using protobuf/text proto?
LLMs already are very familiar with it, there are 100s of protoc plugin to generate code from it, and it's less verbose and more token efficient + testable with protovalidate
Give me a fucking break. You’re treating Blade Runner like some high‑falutin moral fable while the actual underclass is sitting in plain sight, exploited by systems you refuse to even look up. Modern slavery isn’t just a dystopian backdrop; over fifty million people live in some form of forced labor or forced marriage right now, and that number has been climbing, not stabilizing. Tens of millions are stuck in forced labor for companies and states, generating hundreds of billions in illegal profits off human coercion every single year. Countries like India, China, North Korea, Pakistan, Russia, Indonesia, Nigeria, Bangladesh, Turkey, and the United States together account for well over half of all victims, and some of the worst spikes are happening in Iraq, Venezuela, and countries already drowning in climate‑related disasters.
If you want a real starting point that actually hits you in the chest rather than in the feel‑good aesthetic cortex, look at wage theft in the United States. Wage theft outright steals about fifty billion dollars from workers each year. That’s over a hundred times what robberies cost in terms of direct financial loss, and it’s happening to the very people who can least afford it. At least four million workers are paid below the minimum wage illegally every year, each losing roughly three thousand dollars on average, which adds up to over thirteen billion dollars in stolen wages just from that one category. That doesn’t even include unpaid overtime, not being paid for full hours, being misclassified as “independent contractors,” or being forced to work “off the clock” with no record and no recourse.
Low‑wage workers in service jobs, domestic work, farming, and precarious gig‑adjacent roles are the ones getting hammered hardest, with surveys in several cities showing around half of low‑wage workers reporting at least one serious wage violation each year. You can’t get poorer than having time stolen from you, and that’s exactly what’s happening on a massive scale. The feds barely lift a finger: the agency meant to enforce basic wage rules has fewer than six hundred and fifty investigators overseeing more than a hundred and sixty‑five million workers, which is one investigator for roughly three hundred thousand people. In the most recent year they recovered a bit over a quarter billion dollars in back pay, but that’s less than one percent of the estimated fifty billion stolen from workers in that same year. In plain English, the state is letting something like ninety‑five percent or more of wage theft go utterly unpunished.
So no, Blade Runner is not your wake‑up call. The film is a mood board for people who want to pretend they’re deep while keeping their eyes closed to the real indices of exploitation happening right now, right here—the forty million people trapped in modern slavery worldwide, the tens of millions in forced labor, and the fifty billion dollars stolen from workers in your own backyard every year. Those are the numbers that should disgust you, not some neon‑soaked metaphor about replicants. If you can’t feel rage in front of that, you’re not woke; you’re just playing dress‑up in someone else’s hellscape.
---
Fun little project, though!
Please actually attempt to read what I have actually written instead of seeing two words and making grand leaps of assumption.
Keep those two words out of your writing, and I won't read them.
Pay my hourly rate and I'll write whatever you want within reason :)
Not listed in your website AFAICT.
For writing it starts at $200 per hour with a minimum commitment of 10 hours of work.
Cringe.
did you read the post? I think
> I wonder what underclass of people like that exists today.
was a subtle hint to what you are talking about.
I would expect a 1000+ karma account to know better than to shitpost AI slop on hacker news.
(if this is not slop, I don't know what to say... the only way your comment relates to the article is the words "Blade Runner")
You're right, I should do better. This was lazy, but it struck a nerve.